Automated Determination of Topology and Line Parameters in Low Voltage Systems Using Smart Meters Measurements

This work proposes a generic method to utilize customer smart meter measurements to automatically and simultaneously estimate topology, line parameters, and customer and line phasing connections in low voltage (LV) distribution systems. This generic approach is applicable to single, two, and three-phases lines and customers. Hence, it is suitable not only for North American systems but also for European and South American systems. This generic estimation is conducted by using a multiple linear regression model applied to data supplied by customers meters. The acceptance of each estimated parameter is carried out through comparisons with mathematical (e.g., coefficient of determination and relative standard deviation) and physical constraints (e.g., resistances, line length, and conductor X/R ratios). Granularity and sensitivity analyses are also conducted taking into account smart meter data quality (e.g., update ratio, metering errors, resolution, clock desynchronization). The obtained results show the high performance of the method to correctly estimate the system topology, line parameters, and line and customers phasing based on a specification of 15-day sample size with 60-min resolution as a general compromise solution between data acquisition and accuracy.